Can you assign a public IP or URL to an S3 Glacier Vault? I want to use it for automatic backups.
I realize that I can upload to an S3 bucket and then use lifecycle rules to move it over to glacier, but I'm asking if I can skip that step entirely and upload directly to Glacier Vault.
Thanks for any tips!
When originally released, Amazon Glacier was only accessible directly (rather than via Amazon S3). It offers low-cost storage, but it is only accessible via API (not much can be done in the Management Console) and it is very slow because all requests are processed as jobs. This even makes it slow to list the contents of a Vault.
You can certainly access Amazon Glacier directly, but it would be via API calls to the Glacier Endpoint. I would recommend that you use tools such as Cloudberry Backup that know how to talk directly to Glacier.
However, a much simpler way to use Glacier is to store files in Amazon S3 and then select the Glacier or Glacier Deep Archive storage class. This allows use of the S3 interface and the Deep Archive storage class is actually cheaper than Glacier itself! You can also use the AWS CLI to upload backups, which is much easier than working with a Glacier Vault.
By the way, if you are purely wanting to use S3/Glacier for "backups", I would highly recommend using traditional backup tools that know how to use S3. They are much more reliable, and offer more capabilities, than doing it yourself. For example, they can keep multiple versions of files and can retain deleted files for a period of time to allow recovery.